While genomic programs provide ever more sophisticated information on the sequence and patterns of expression of mammalian genes, it is increasingly recognized that integrating this information into a functional model of how a cell works requires an understanding of how the protein products of expressed genes interact within the cell. Although we have made significant improvements in our ability to clone, sequence and analyze DNA sequences, our reciprocal abilities for studying RNA and protein molecules are significantly less facile or advanced. Furthermore, proteins themselves represent significantly more complex molecules in terms of composition, shape and activity compared to double stranded DNA. A central challenge facing workers in the field today is to understand out how a protein's activity and function within a cell are regulated and coordinated within the native physiological context.
Traditionally, genetic analysis has been used for determining the function of gene products and how they interact with other proteins within a common pathway. Unfortunately genetic analysis in vertebrate organisms is extremely time consuming and expensive. An alternative approach is to devise an assay system for a given protein and then screen for compounds that activate or inhibit its function. These compounds can be used to dissect the cellular pathways the protein functions in, as well as serving as potential compounds of therapeutic value.
Although there is tremendous interest in understanding the regulation and interactions of proteins within cells there are relatively few methods that are robust, simple to use, amenable to high throughput screening or can be used effectively within living cells. Furthermore in many cases where specific assays do exist these are restricted in scope to individual enzymatic steps or to one or two defined pathways.
A need thus exists for sensitive methods of interfacing the functional modifications of proteins with optical signals that can be used detect and monitor these changes, for example for use in high throughput screening. In drug screening applications these assays can be applied to find useful compounds that are specific and selective for a particular protein or signal transduction or metabolic pathway.
Proteins may undergo a huge variety of post-translational modifications subsequent to their synthesis in the cell. In many cases these modifications can play critical roles in the functioning and stability of the modified proteins. For example, proteolysis, phosphorylation, covalent attachment of a lipid or lipid derivative, disulfide bond formation, glycosylation and oxidation all can have important functional effects. Many other examples also exist and may play important functional roles within a cell for defined proteins.
One approach to developing a generic assay capable of detecting these myriad post-translational modifications is to operatively couple these activities through a central pathway of protein modification that can be sensitively measured with a common reporter system. In the present invention, the inventors have recognized that by coupling post-translational activities to the stability of a high sensitivity reporter moiety it is possible to develop uniform cell based assays for a range of activities. Importantly these measurements are robust enough for high throughput screening applications, readily adaptable to a range of activities and provide cellular assays that provide information within a living cell.
In the present invention, post-translational activities can be measured by providing one or more constructs in which the activity to be measured influences the stability of a reporter moiety. In one embodiment, this may be achieved by providing a reporter moiety that is operatively coupled to a multimerized destabilization domain via a linking moiety. The linking moiety comprises a recognition motif for the target activity such that modification of the linker by the activity results in altered stability of the reporter moiety. If the reporter moiety is an enzymatic reporter gene the method provides a high sensitivity readout that is generally applicable to a range of activities which are otherwise difficult to measure within living cells. The multimerized destabilization domain described herein provides a key advantage in the method because it enables the degree of destabilization to be predictably tuned to any activity level or intrinsic stability of the target protein or reporter moiety.
The regulation of protein stability is an area of particular interest because of its increased recognition as a key regulator of a protein's concentration and function in the cell. Although our knowledge of the factors that control protein stability have grown dramatically in recent years, it is clear that a variety of cellular pathways and environmental cues participate in and control a protein's fate. For example, mis-folding, proteolysis, oxidation and some conformational changes that expose significant surface hydrophobicity readily contribute to the recognition of a protein by the cellular machinery for protein degradation. The majority of cytoplasmic protein degradation involves the ubiquitination of the target protein followed by binding and degradation by the proteasome. (For review see Hershko and Ciechanover (1998) Annu. Rev. Biochem. 67 425-79)
A key step in protein ubiquitination, and degradation, is recognition of the target protein by ubiquitin protein ligase or E3 enzyme. This class of enzymes is responsible for the covalent attachment of ubiquitin to the target protein via an amide isopeptide linkage to an ε-amino group of one of the substrate protein's lysine residues. There are currently believed to be multiple families of E3 enzymes, additionally there is increasing evidence that some E3 proteins exist as multi-subunit protein complexes (Laney and Hochstrasser (1999) Cell 97 427-430). E3 proteins and their associated complexes are believed to be largely responsible for recognizing and ubiquitinating damaged proteins as well as specific destabilization domains present in target proteins. Once recognized, a protein target that has been modified by the addition of a single ubiquitin domain, becomes a substrate for further ubiquitination, either at different sites in the substrate protein, or through extension of the conjugated ubiquitin. This process can thus lead to a poly-ubiquitinated protein with numerous branched ubiquitin domains attached. Once poly-ubiquitinated, the protein is recognized with high affinity by the proteasome where it is degraded.
The addition of specific destabilization domains to a target protein has in some cases been demonstrated to destabilize that target protein. A key challenge in this area has been to provide a predictable way of creating graded levels of destabilization for a given protein that that can be utilized in manipulating the steady state levels or dynamic temporal regulation of that protein. The present inventors have discovered for the first time that by providing stable multimerized linear chains of individual destabilization domains, such as ubiquitin, it is possible to create a generic method of protein destabilization that is widely applicable to virtually any protein. Importantly, this approach has the advantage that the degree of destabilization can be accurately controlled by varying the number of destabilization domains added to the target protein. As a result, the actual cellular concentration and half-life of an exogenously expressed protein in a cell or living organism can be accurately and reproducibly controlled. By coupling 1, 2, 3, 4 or more copies of ubiquitin to the reporter gene β-lactamase it has been possible to regulate the protein concentration of this protein in the cell over a 10-fold range compared to the native protein. The present inventors have applied this discovery to create an assay technology that is broadly capable of measuring a wide range of post-translational activities.